In recent years, there has been an increasing interest in the use of ceramics for structural applications historically served by metals. The impetus for this interest has been the superiority of ceramics with respect to certain properties, such as corrosion resistance, hardness, wear resistance, modulus of elasticity, and refractory capabilities when compared with metals.
However, a major limitation on the use of ceramics for such purposes is the feasibility and cost of producing the desired ceramic structures. For example, the production of ceramic boride bodies by the methods of hot pressing, reaction sintering and reaction hot pressing is well known. In the case of hot pressing, fine powder particles of the desired boride are compacted at high temperatures and pressures. Reaction hot pressing involves, for example, compacting at elevated temperatures and pressures boron or a metal boride with a suitable metal-containing powder. U.S. Pat. No. 3,937,619 to Clougherty describes the preparation of a boride body by hot pressing a mixture of powdered metal with a powdered diboride, and U.S. Pat. No. 4,512,946 to Brun describes hot pressing ceramic powder with boron and a metal hydride to form a boride composite.
However, these hot pressing methods require special handling and expensive special equipment, they are limited as to the size and shape of the ceramic part produced, and they typically involve low process productivities and high manufacturing cost.
A second major limitation on the use of ceramics for structural applications is their general lack of toughness (i.e. damage tolerance or resistance to fracture). This characteristic tends to result in sudden, easily induced, catastrophic failure of ceramics in applications involving even rather moderate tensile stresses. This lack of toughness tends to be particularly common in monolithic ceramic boride bodies.
One approach to overcome this problem has been to attempt to use ceramics in combination with metals, for example, as cermets or metal matrix composites. The objective of this approach is to obtain a combination of the best properties of the ceramic (e.g. hardness and/or stiffness) and the metal (e.g. ductility). U.S. Pat. No. 4,585,618 to Fresnel, et al., discloses a method of producing a cermet whereby a bulk reaction mixture of particulate reactants, which react to produce a sintered self-sustaining ceramic body, is reacted while in contact with a molten metal. The molten metal infiltrates at least a portion of the resulting ceramic body. Exemplary of such a reaction mixture is one containing titanium, aluminum and boron oxide (all in particulate form), which is heated while in contact with a pool of molten aluminum. The reaction mixture reacts to form titanium diboride and alumina as the ceramic phase, which is infiltrated by the molten aluminum. Thus, this method uses the aluminum in the reaction mixture principally as a reducing agent. Further, the external pool of molten aluminum is not being used as a source of precursor metal for a boride forming reaction, but rather it is being utilized as a means to fill the pores in the resulting ceramic structure. This creates cermets which are wettable and resistant to molten aluminum. These cermets are particularly useful in aluminum production cells as components which contact the molten aluminum produced but preferably remain out of contact with the molten cryolite.
European Application 0,113,249 to Reeve, et al. discloses a method for making a cermet by first forming in situ dispersed particles of a ceramic phase in a molten metal phase, and then maintaining this molten condition for a time sufficient to effect formation of an intergrown ceramic network. Formation of the ceramic phase is illustrated by reacting a titanium salt with a boron salt in a molten metal such as aluminum. A ceramic boride is developed in situ and becomes an intergrown network. There is, however, no infiltration, and further the boride is formed as a precipitate in the molten metal. Both examples in the application expressly state that no grains were formed of TiAl.sub.3, AlB.sub.2, or AlB.sub.12, but rather TiB.sub.2 is formed demonstrating the fact that the aluminum is not the metal precursor to the boride.
U.S. Pat. No. 3,864,154 to Gazza, et al. discloses a ceramic-metal system produced by infiltration. An AlB.sub.12 compact was impregnated with molten aluminum under vacuum to yield a system of these components. Other materials prepared included SiB.sub.6 -Al, B-Al; B.sub.4 C-Al/Si; and AlB.sub.12 -B-Al. There is no suggestion whatsoever of a reaction, and no suggestion of making composites involving a reaction with the infiltrating metal nor of any reaction product embedding an inert filler or being part of a composite.
U.S. Pat. No. 4,605,440 to Halverson, et al., discloses that in order to obtain B.sub.4 C-Al composites, a B.sub.4 C-Al compact (formed by cold pressing a homogeneous mixture of B.sub.4 C and Al powders) is subjected to sintering in either a vacuum or an argon atmosphere. There is no infiltration of molten metal from a pool or body of molten precursor metal into a preform. Further, there is no mention of a reaction product embedding an inert filler in order to obtain composites utilizing the favorable properties of the filler.
While these concepts for producing cermet materials have in some cases produced promising results, there is a general need for more effective and economical methods to prepare boride-containing materials.